The in-depth exploration and development of knowledge organization methodologies tailored to the unique characteristics of traditional Chinese medicine (TCM) has become a pivotal issue in advancing the modernization of TCM knowledge systems and fostering scientific innovation. Through systematic review and analysis of existing TCM knowledge organization research, this study identifies current limitations. It proposes exploring integrated approaches for integrating ancient and contemporary TCM knowledge within existing classical text repositories, constructing theoretical knowledge bases, conducting ontological mapping studies between historical and modern TCM knowledge, and developing efficient human-machine collaborative annotation methods to build knowledge graphs. These initiatives aim to provide actionable strategies for optimizing TCM knowledge organization frameworks.

Objective By applying data mining technology, this study collects, organizes, and researches the ancient prescriptions for treating palpitations in the “Chinese Formula Database”, analyzes their medication rules, and provides references for modern clinical medication and new drug research and development. Methods First, the medical records related to palpitations in the database were screened, and a Microsoft Excel dataset was created. Then, the”Ancient and Modern Medical Case Cloud Platform” was used to conduct frequency statistics on the drugs, as well as their properties, flavors, meridian tropisms and efficacies. Finally, IBM SPSS Modeler 18.0 and IBM SPSS Statistics 25 data-processing software were applied to perform association rule analysis and cluster analysis on high-frequency drugs (frequency$>100$, rate$> 0.85 \%$), so as to explore the compatibility rules among them. Results A total of$\mathbf{1, 1 5 7}$prescriptions were included, involving 639 kinds of drugs. Renshen had the highest frequency of use; drugs with warm property and sweet flavor were used most frequently; drugs that enter the spleen meridian were the most common; and drugs with the effect of calming the mind and improving intelligence had the highest frequency of use. Through association rule analysis and cluster analysis,$\mathbf{1 5}$groups of core drug pairs and 4 groups of core drug combinations were obtained. Conclusion Most ancient prescriptions for treating palpitations adopt four therapeutic approaches: promoting fluid production and nourishing blood, replenishing qi and consolidating the root, simultaneous treatment of the heart and spleen, and drying dampness and resolving phlegm. These approaches accurately correspond to the core pathogeneses of palpitations, namely, malnutrition of the heart spirit, deficiency of qi and blood, and phlegm stasis disturbing the heart.

Objective To develop a unified and standardised metadata framework for ancient books of traditional Chinese medicine, aiming to achieve standardised and regulated management of such resources. This initiative seeks to enhance the efficiency of managing and utilising ancient books of traditional Chinese medicine while providing a scientific guideline for their description. Methods Using literature review, comparative analysis, and expert consultation methods, this study analyzes and compares the current state of ancient books of traditional Chinese medicine metadata and relevant domestic and international standards. Based on the Dublin Core Element Set (DC) as the basic framework and tailored to the characteristics of ancient books of traditional Chinese medicine, it proposes a research idea and framework for ancient books of traditional Chinese medicine metadata standards. Results This study proposes the research ideas and principles of metadata standards for ancient Chinese medicine books centered on the knowledge system of traditional Chinese medicine, clearly stipulates the objects and information sources of ancient Chinese medicine book resources, constructs a metadata structure model and hierarchical relationship including core metadata, extended metadata and element modifiers, and defines the core metadata elements. Conclusion The metadata standard for ancient books of traditional Chinese medicine will facilitate the integration and sharing of ancient books of traditional Chinese medicine resources, thereby advancing the inheritance and development of traditional Chinese medicine knowledge.

Objective To systematically evaluate and analyze the prescriptions for treating Xiongbi (Chest Bi-Syndrome) in ancient Traditional Chinese Medicine (TCM) books by integrating evidence-based valuation of ancient TCM literature and data mining methods. Methods First, the screened vidence of TCM prescriptions for Xiongbi from ancient books was evaluated and graded using the Evaluation and Grading Scale for Evidence of Disease Prevention and Treatment in Ancient TCM Books. For high-level evidence, the “Ancient and Modern Medical Records Cloud Platform” was applied to analyze indicators including drug use frequency, efficacy categories, nature, taste, and meridian tropism of drugs. Additionally, association rule analysis and cluster analysis were conducted. Results A total of 155 pieces of evidence from ancient TCM books for treating Xiongbi were finally included. Through evaluation, 31 prescriptions (e.g., Renshen Decoction, Fuling Xingren Gancao Decoction, Gualou Xiebai Banxia Decoction) were identified as high-level evidence. Data mining results showed that, among the high-level evidence for Xiongbi treatment in ancient books, the main drug efficacies were dispelling cold to relieve pain, tonifying fire to support yang, and warming and unblocking meridians; warm-natured drugs were most frequently used; pungent was the dominant taste of drugs; and the spleen meridian was the primary meridian tropism. Association rule analysis revealed that Cinnamomi Cortex (cinnamon) was a common drug for treating Xiongbi with heart pain, and the “Ginseng-Cinnamomi Cortex” pair was the most frequently used drug combination for Xiongbi. Cluster analysis yielded 4 core prescription groups, each reflecting distinct treatment strategies. Conclusion This study conducts quality evaluation and in-depth mining analysis of prescriptions for Xiongbi in ancient TCM books using evidence-based evaluation of ancient TCM literature and data mining. The results not only provide more reliable evidence from ancient books for the clinical treatment of Xiongbi in TCM but also offer sufficient ancient book-based evidence support for the formulation of TCM clinical decisions.

Objective: By comparing the medical cases of the renowned traditional Chinese medicine (TCM) practitioner Yu Ying-ao with the original patterns of primary epilepsy in ancient medical records, this study aims to explore reference pathways for the inheritance and innovation of the experiences of renowned TCM practitioners. Method: The Jaccard similarity comparison method is utilized for multi-dimensional comparison and analysis. This involves comparing the associations between syndromes and medications, treatment principles, collections of Chinese medicines, categorization of Chinese medicine effects, and similarities and differences in meridian tropisms in the medical cases of Yu Ying-ao with those of ancient primary epilepsy treatments. Result: In terms of treatment, Yu Ying-ao inherited the basic principle of “resolving phlegm and calming” in ancient medical records, the overall approach to treatment should focus on suppressing seizures and stopping epilepsy, as well as transforming phlegm and unblocking collaterals. The specific treatment varies depending on the condition, with options such as calming the spirit, strengthening the spleen, adjusting the liver, moisturizing the intestines, nvigorating spleen- stomach, and replenishing qi. In terms of core TCM formulas, a comparison with ancient medical records reveals that the essential TCM combinations for the syndrome of phlegm- heat disturbing the spirit, as identified by Yu Ying-ao, are more refined. Both formulas include (Dannanxing), (Baifu), (Yujin), (Danshen), And (Zhuru). In terms of the scope of medication, the frequently used drugs in ancient medical records are mostly those that are highly used in clinical practice by Yu Ying-ao. In terms of efficacy, there has been a high degree of continuity, with both ancient and modern times valuing the use of phlegm- eliminating, heat- clearing, calming, qi regulating, and tonifying vacuity medications. However, Yu Ying-ao tends to emphasize the use of Activating blood stasis and suppressing liver- yang medications in his clinical practice. From the perspective of TCM meridian tropism, ancient medical records most frequently mentioned lung meridian as the primary meridian for TCM prescriptions, followed by spleen, heart, and liver meridians. In the medical cases of Yu Ying-ao, liver meridian was the most frequently mentioned as the meridian for TCM prescriptions, followed by lung, heart, and spleen meridians. Conclusion: Through multidimensional comparative analysis of the composition of core TCM medicines, treatment principles and methods, collections of TCM medicines, categorization of TCM medicine effects, and meridian tropism, reference pathways can be provided for the inheritance and innovation of experienced TCM practitioners.

Although the Conformer model excels in speech processing, its core self-attention mechanism is limited in capturing multi-scale temporal dynamics and lacks explicit modeling of frequency-domain features, both crucial for Speech Emotion Recognition (SER). To address this, we propose ConMSDMamba, a novel Conformer-based architecture for SER. Specifically, to overcome the single-scale limitation of the original self-attention, we introduce a multi-scale dilated structure with parallel dilated convolutions to capture diverse temporal contexts. We further find that combining this structure with bidirectional Mamba models long-range temporal dependencies more efficiently than multi-head self-attention. Furthermore, to complement the Conformer's time-domain focus, we design a time-frequency convolution module that incorporates a wavelet-based branch for joint time-frequency perception. Experimental results on the widely used IEMOCAP and MELD datasets demonstrate that ConMSDMamba outperforms state-of-the-art methods.

Objective To keep up with the progress of treatment for myopia in traditional Chinese medicine (TCM) through literature research, so as to determine data source of TCM diagnosis and treatment for myopia, from which entities are extracted for integration of relevant knowledge, and further formal expression and visualization of knowledge. Methods Literature research and comparative analysis were used to determine the data source, and the seven-step method of constructing ontology was referred to for extraction of TCM diagnosis and treatment entities, which then was terminalized for development of classes and class hierarchy, and class properties, as well as creation of instances before consistency test of the ontology. Results The ontology of TCM diagnosis and treatment for myopia includes 26 classes and 124 instances, 9 object properties and 1 data property. Conclusion: The combination of ontology technology and TCM diagnosis and treatment knowledge of myopia contributes to the construction of knowledge base of TCM diagnosis and treatment ontology for myopia, which realizes formal expression and visualization of relevant TCM knowledge, lays a foundation for the construction of TCM top-level ontology and ophthalmology ontology, and facilitates knowledge reusing and sharing.

Objective To provide theoretical basis and data support for the treatment of Xiongbi by analyzing the compatibility and application rules of treating Xiongbi in traditional Chinese medicines (TCMs) based on data mining. Methods Chinese herbal formulas for Xiongbi were searched in the "Chinese formula database" and collected. The usage frequency, efficacy category, and nature and flavour attribution of each medicine in the prescriptions were statistically analyzed. The association rules and clustering analysis were conducted for those Chinese medicines with a usage frequency above 10% by using the IBM SPSS Modeler 18.0 and SPSS Statistics 27 data-processing software. Results A total of 455 prescriptions for Xiongbi in TCM classics were obtained in this study, involving 324 flavors of Chinese medicines, 15 of them had a frequency of use above 10%. The top three medicine ranked according to the frequency were Rougui (170 times), Gancao (128 times), and Danggui (121 times).The efficacy of the drug is based on dispersing cold and relieving pain, tonifying fire to help Yang, and moistening the bowels to relieve constipation;In addition, our results indicated that the main medicine treating Xiongbi belongs to the mild on medicinal property, which were pungent taste and entering the spleen meridian. Association rule analysis showed that Rougui was the core medicine, and the compatibility of "Ganjiang, Rougui" was the core drug pair of medicines for the treatment of Xiongbi. Three potential new prescriptions were obtained from the cluster analysis, each reflecting a different therapeutic idea. Conclusion Rougui was the core medicine in the Xiongbi treatment, the core drug pair of medicines was "Ganjiang, Rougui". Accordingly, Tongyang Sanjie, Zaoshihuatan and Xinpitongzhi are the core ideas of Chinese medicine in treating Xiongbi.

To analyze the discrimination and treatment of "Cold-Dampness Depression Lung Syndrome" of COVID-19 in TCM classics. Methods: Using the mathematical statistics and data mining methods to sort and analyze information of prescriptions treating "Cold-Dampness Depression Lung Syndrome" of COVID-19 in TCM classics. Results: 50 ancient prescriptions with therapeutic effects were selected, contain contain 125 traditional Chinese medicines, and the top 5 are Gancao (Glycyrrhizae Radix Rhizoma), Banxia(Pinelliae Rhizoma), Renshen (Ginseng Radix Et Rhizoma), Baizhu(Atractylodis Macrocephalae Rhizoma) and Chenpi(Citri Reticulatae Pericarpium) in order of frequency of use. The meridians of the medicines are mainly lung meridian, spleen meridian and stomach meridian, and the properties of the medicines are mostly warm, followed by mlid and lukewarm. The main medicinal pairs are Jiegeng (platycodonis Radix)-Gancao (Glycyrrhizae Radix Rhizoma), Baishao (Paeoniae Radix Alba)-Gancao (Glycyrrhizae Radix Rhizoma),Mahuang (Ephedrae Herba)-Gancao (Glycyrrhizae Radix Rhizoma),Chuanxiong (Chuanxiong Rhizoma)- Gancao (Glycyrrhizae Radix Rhizoma) and Cangnzhu (Atractylodis Rhizoma)-Gancao (Glycyrrhizae Radix Rhizoma).Conclusion: By analysing the ancient prescriptions with potential treatment for "Cold-Dampness Depression Lung Syndrome" of COVID-19, we found high-frequency medicines and medicinal pairs, and had a more comprehensive understanding of the treatment of COVID-19, can provide a reference for the research of COVID-19 specific medicines.

Objective To address the issues in named entity recognition (NER) in the field of traditional Chinese medicine (TCM), this study proposes a method for identifying entities in TCM herbal literature; Methods We identify and describe the types of knowledge entities and entity relationships involved in herbal literature. We apply the BIO sequence labeling method to generate a training corpus dataset and use our self-developed CNLP text annotation system for text annotation. The Bert model is employed for recognizing named entities; Results The Bert model achieved entity recognition results for various entities in TCM herbal literature with precision (P) of 71.49%, recall (R) of 72.33%, and F1 score of 71.91%; Conclusion The Bert model demonstrates a certain level of applicability in recognizing various entities in TCM herbal literature. This model is helpful in extracting valuable structured information from a large volume of text data.

Objective Using literature data mining methods, we analyzed the frequency of medication, combinations of drugs, and so on in the treatment of goiter disease with ancient classic prescriptions. We aimed to uncover the core combinations and new prescriptions for treating goiter; s We searched the "Ancient Classic Prescription Database" for literature related to the treatment of goiter, selected the prescriptions for treating goiter, entered them into the medical case cloud platform of ancient and modern times, and used rule analysis, cluster analysis, and other data mining methods to analyze the rules of prescription combination. Results The study finally included 145 prescriptions. The results of drug frequency statistics showed that the use frequency of kelp, seaweed, and Pinellia was relatively high. The analysis of Chinese medicine properties revealed that cold and warm medicines were used frequently, as were bitter, pungent, and salty medicines. Medicines entering the stomach, liver, and kidney meridians were used the most. The results of association rule analysis and cluster analysis revealed the combination relationship and classification of Chinese medicines. Complex network analysis identified the core prescription composition for treating goiter in ancient classic prescriptions, including kelp, seaweed, and Pinellia. Conclusion: This study analyzed the treatment of goiter with ancient classic prescriptions through data mining and found that kelp, seaweed, and Pinellia might be the core combination for treating goiter. The research results can provide a reference for the clinical practice of traditional Chinese medicine in treating goiter.

Objective: To analyze the differentiation and treatment principles of Swollen-head Infection in TCM classics. Methods: Ancient medical case data related to warm diseases were selected as the data source, and the standard principles of data extraction were formulated. Data mining methods such as mathematical statistics, factor analysis, cluster analysis, and association rules were used to systematically sort out and analyze the etiology, location, syndrome, treatment, formulations and other information of Swollen-head Infection. Results: Swollen-head Infection is primarily attributed to pathogenic wind and heat toxins. The significance of "Li Qi" (Epidemic pathogen) should be emphasized.The disease primarily affects the head, and the pathogenic factors tend to linger in the lung-defense. The clinical manifestations are closely related to the affected area of the head, often accompanied by other systemic symptoms. The treatment approach commonly involves combining internal and external therapies. Combinations of herbs such as Xuanshen (Scrophulariae Radix)-Lianqiao (Forsythiae Fructus), Xuanshen (Scrophulariae Radix)-Huangqin(Scutellariae Radix), Jiegeng(Platycodonis Radix)-Lianqiao (Forsythiae Fructus), Jiegeng(Platycodonis Radix)-Huangqin(Scutellariae Radix), Chaihu(Bupleuri Radix)- Jiegeng(Platycodonis Radix), and Chaihu(Bupleuri Radix)- Huanglian(Coptidis Rhizoma) are notable for their abilities to clear heat, detoxify, disperse wind, and eliminate pathogenic factors. Additionally, Puji Xiaodu Yin and its modifications are considered essential medications for treating Swollen-head Infection. Conclusion: Through the data mining of the rules of syndrome and prescription of Swollen-head Infection in ancient books of warm diseases, to provide reference for the differentiation and treatment of head and face swelling and poison infectious diseases.

Traditional Chinese medicine (TCM) has accumulated thousands years of knowledge in herbal therapy, but the use of herbal formulas is still characterized by reliance on personal experience. Due to the complex mechanism of herbal actions, it is challenging to discover effective herbal formulas for diseases by integrating the traditional experiences and modern pharmacological mechanisms of multi-target interactions. In this study, we propose a herbal formula prediction approach (TCMFP) combined therapy experience of TCM, artificial intelligence and network science algorithms to screen optimal herbal formula for diseases efficiently, which integrates a herb score (Hscore) based on the importance of network targets, a pair score (Pscore) based on empirical learning and herbal formula predictive score (FmapScore) based on intelligent optimization and genetic algorithm. The validity of Hscore, Pscore and FmapScore was verified by functional similarity and network topological evaluation. Moreover, TCMFP was used successfully to generate herbal formulae for three diseases, i.e. the Alzheimer's disease, asthma and atherosclerosis. Functional enrichment and network analysis indicates the efficacy of targets for the predicted optimal herbal formula. The proposed TCMFP may provides a new strategy for the optimization of herbal formula, TCM herbs therapy and drug development.

The early stage of brucellosis belongs to the category of latent temperature disease in traditional Chinese medicine, and the symptom is damp-heat epidemic pathogen. But later to joint pain, soreness and weakness of waist and knees, fatigue, joint pain, night sweats, etc., mainly in the joint, belongs to the liver and kidney deficiency, qi and blood deficiency, spleen wet turbidity syndrome, belongs to the category of traditional Chinese medicine rheumatism heat arthralgia syndrome, etiology and pathogenesis due to ' winter does not store essence, spring must disease temperature ' within the virtual cause. Based on the syndrome differentiation of traditional Chinese medicine, Duhuo Jisheng Decoction combined with Simiao Pill was used to treat liver and kidney, nourish qi and blood, and eliminate dampness and turbidity, so as to achieve the purpose of strengthening the body and eliminating evil, restoring healthy qi and eliminating evil qi. To provide reference for the treatment of infectious diseases such as brucellosis.

Objective The prescription rules of traditional Chinese medicine (TCM) for knee osteoarthritis (KOA) in ancient medical classics were explored based on Traditional Chinese Medicine Inheritance Computer System (TCMICS) to provide reference and evidence for modern clinical treatment. Methods The contents of TCM treatment for KOA in ancient medical classics were comprehensively collected and then logged in TCMICS after screening, where frequency of formulae, medicinals and medicinal combination, as well as properties, flavors and channel tropism of frequently-used medicinals, etc. were counted. Results A total of 510 items were selected from ancient medical classics, including 221 formulae with specific names, and 25 medicinals with a frequency of more than 40. The top three formulae which were most frequently used included Da Fangfeng Tang (Major Ledebouriella Decoction), Wutou Tang (Aconite Decoction), and Duhuo Jisheng Tang (Pubescent Angelica and Mistletoe Decoction). The top five medicinals which were most frequently used included Danggui (Angelicae Sinensis Radix), Niuxi (Achyranthis Bidentatae Radix), Fangfeng (Saposhnikoviae Radix), Chuanxiong (Chuanxiong Rhizoma), and Gancao (Glycyrrhizae Radix et Rhizoma). It was also found through analysis that the core formula was modification of Da Fangfeng Tang. Conclusion The main formulae used for KOA are Da Fangfeng Tang, Wutou Tang, and Duhuo Jisheng Tang, while the selection of medicinals highlights the actions of nourishing the liver and kidney, boosting qi and blood, or activating blood and removing blood stasis, and warming the channels and dissipating cold, which can provide reference for clinical treatment.

Detecting out-of-domain (OOD) intents is critical in a task-oriented dialog system. Existing methods rely heavily on extensive manually labeled OOD samples and lack robustness. In this paper, we propose an efficient adversarial attack mechanism to augment hard OOD samples and design a novel generative distance-based classifier to detect OOD samples instead of a traditional threshold-based discriminator classifier. Experiments on two public benchmark datasets show that our method can consistently outperform the baselines with a statistically significant margin.

The performance of many intent detection approaches will degrade when they meet open-set data because there is out-of-domain (OOD) noise. Some works utilize clean but expensive labeled data to supervise models more robust to the varied environment. However, a large number of labeled samples are scarce and their models no longer change after finishing training. To address this problem, we propose an iterative learning framework that can dynamically improve the model's ability of OOD intent detection and meanwhile continually obtain valuable new data to learn deep discriminative features. Concretely, the model can generate pseudo labels for unlabeled examples by self-training and use the local outlier factor (LOF) algorithm to detect unknown intents. Furthermore, we add mask operation on supervised contrastive loss (SCL) and use the masked-SCL to absorb new data effectively. Experiments on CLINC and SNIPS demonstrate that our proposed method can robustly realize intent detection in the presence of a high proportion of open-set.

Objective: Based on the integrated pharmacology platform of Chinese medicine V2.0 (TCMIP V2.0), this paper explores the molecular mechanism and quality markers of Tanyu Tongzhi Recipe against myocardial ischemia reperfusion injury (MIRI). Methods: Use TCMIP V2.0 to collect the drug components, targets, and MIRI disease targets of the recipe for phlegm and blood stasis, construct a drug-disease-target interaction network, screen drugs and disease common targets, analyze the biological process of the common targets, and finally establish a “component-target-pathway-pharmacological action” multi-dimensional network analysis and analysis of the core components in the network. Results: A total of 157 medicinal chemical components, such as Pinellia, Red Peony, Chuanxiong, Licorice, etc. were collected from the prescription of phlegm and blood stasis, with 272 corresponding targets. A total of 290 MIRI disease targets were collected. Through "Traditional Chinese Medicine Association Network Mining," the first 100 targets in the core targets were analyzed and 31 targets shared by drugs and diseases. Analysis of these shared targets revealed that the shared targets are mainly involved in the positive regulation of tube formation, angiogenesis, positive regulation of autophagy, inflammation, and negative regulation of apoptosis. KEGG enrichment analysis shows that these common targets are mainly enriched in the FoxO signaling pathway, HIF-1 signaling pathway, TNF signaling pathway, PI3K-Akt signaling pathway, VEGF signaling pathways and other signal pathways. Further analysis of the multi-dimensional network revealed that 46 components of Tanyu Tongzhi Prescription affect the above five signal pathways to exert anti-MIRI effects by interacting with 19 shared targets. Among them, quercetin, kaempferol, naringenin, and baicalein are intervened. AKT affects five signal pathways involved in the occurrence and development of MIRI. Conclusion: Quercetin, kaempferol, naringenin, and baicalein in Tanyu Tongzhi Recipe can resist MIRI by regulating inflammation, autophagy and apoptosis, oxidative stress, angiogenesis, and other processes.

Objective: To analyze medication rule of the treatment for spring warm disease in case records of Qing Dynasty (1636-1912) based on data mining. Methods: Ancient case records of warm disease were selected as the data source and the standard principle of data extraction was established. TCM Miner was used for collecting data and analyzing association rules. Results: A total of 225 case records of spring warm disease are recorded in 20 ancient medical classics, including 37 formulas and 219 Chinese medicines. The main formulas include Baihu Tang (White Tiger Decoction), Fumai Tang (Pulse-Restorative Decoction) and Liuyi San (Six-to-One Powder), which fall into the category of heat-clearing formula, tonifying formula and phlegm-expelling formula respectively. The main Chinese medicines include Lianqiao (Fructus Forsythiae), Fuling (Poria) and Shichangpu (Rhizoma Acori Tatarinowii), which are capable of relieving cough and panting and resolving phlegm, promoting urination and eliminating dampness, as well as tonifying. The properties are mainly warm, cold and neutral; the flavors are mainly sweet, bitter and acrid; and the medicines mainly enter meridians of the lung, stomach and liver respectively. The frequently-used medicinal combinations include Xingren ((Armeniacae Amarum)) -Beimu (Bulbus Fritillariae), Jupi (Citri Exocarpium)-Xingren, Lianqiao-Jinyinhua (Flos Lonicerae Japonicae) -Xuanshen (Radix Scrophulariae), and Lianqiao-Xuanshen-Shichangpu. Conclusion: The formulas and medications in the treatment of spring warm disease in ancient case records focus on clearing heat that is assisted by eliminating exterior pathogen and protecting yin fluid.

Detecting out-of-domain (OOD) input intents is critical in the task-oriented dialog system. Different from most existing methods that rely heavily on manually labeled OOD samples, we focus on the unsupervised OOD detection scenario where there are no labeled OOD samples except for labeled in-domain data. In this paper, we propose a simple but strong generative distance-based classifier to detect OOD samples. We estimate the class-conditional distribution on feature spaces of DNNs via Gaussian discriminant analysis (GDA) to avoid over-confidence problems. And we use two distance functions, Euclidean and Mahalanobis distances, to measure the confidence score of whether a test sample belongs to OOD. Experiments on four benchmark datasets show that our method can consistently outperform the baselines.

Open-vocabulary slots, such as file name, album name, or schedule title, significantly degrade the performance of neural-based slot filling models since these slots can take on values from a virtually unlimited set and have no semantic restriction nor a length limit.In this paper, we propose a robust adversarial model-agnostic slot filling method that explicitly decouples local semantics inherent in open-vocabulary slot words from the global context.We aim to depart entangled contextual semantics and focus more on the holistic context at the level of the whole sentence.Experiments on two public datasets show that our method consistently outperforms other methods with a statistically significant margin on all the open-vocabulary slots without deteriorating the performance of normal slots. *The first two authors contribute equally.Weiran Xu is the corresponding author.

In task-oriented dialogue systems, dialogue state tracking (DST) is an essential part which aims to estimate user goal at every turn 1 of the dialogue. At each turn, DST aims to estimate user goals by current user utterance and last system action. However, most current approaches encode these relevant sequences by recurrent networks, which is a challenge for these models to capture long-range dependencies. Besides predicting the current dialogue state is relevant to historical context and sometimes refers to the past utterances. In this paper, we propose the Gated Attentive Convolutional network Dialogue State Tracker (GAC) which overcomes these challenges by utilizing the gated attentive convolutional encoder and introducing historical information. We apply a gated attentive convolutional network encoder to learn the sequence representation and introduce historical dialogue utterances as evidence to track the user goal. Experiments show that our method improves joint goal accuracy on WoZ 2.0[1] and MultiWoZ 2.0[2] datasets respectively outperforms the previous state-of-the-art GLAD [3] and Trade [4] model1.

Sequence labeling is a fundamental task of natural language understanding. Recent neural models for sequence labeling task achieve significant success with the availability of sufficient training data. However, in practical scenarios, entity types to be annotated even in the same domain are continuously evolving. To transfer knowledge from the source model pre-trained on previously annotated data, we propose an approach which learns label-relational output structure to explicitly capturing label correlations in the latent space. Additionally, we construct the target-to-source interaction between the source model MSand the target model MTand apply a gate mechanism to control how much information in MSand MTshould be passed down. Experiments show that our method consistently outperforms the state-of-the-art methods with a statistically significant margin and effectively facilitates to recognize rare new entities in the target data especially.